Even at the present time when medical science has advanced, the 5-year survival rate of cancer patients, particularly solid tumor patients (other than blood cancer patients) is less than 50%, and about ⅔ of all cancer patients are diagnosed at an advanced stage and almost all die within 2 years after cancer diagnosis. Such poor results in cancer therapy are not only the problem of therapeutic methods, but also due to the fact that it not easy to diagnose cancer at an early stage and to accurately diagnose advanced cancer and to carry out the follow-up of cancer patients after cancer therapy.
In current clinical practice, the diagnosis of cancer is confirmed by performing tissue biopsy after history taking, physical examination and clinical assessment, followed by radiographic testing and endoscopy if cancer is suspected. However, the diagnosis of cancer by the existing clinical practices is possible only when the number of cancer cells is more than a billion and the diameter of cancer is more than 1 cm. In this case, the cancer cells already have metastatic ability, and at least half thereof have already metastasized. Meanwhile, tumor markers for monitoring substances that are directly or indirectly produced from cancers are used in cancer screening, but they cause confusion due to limitations in accuracy, since up to about half thereof appear normal even in the presence of cancer, and they often appear positive even in the absence of cancer. Furthermore, the anticancer agents that are mainly used in cancer therapy have the problem that they show an effect only when the volume of cancer is small.
The reason why the diagnosis and treatment of cancer are difficult is that cancer cells are highly complex and variable. Cancer cells grow excessively and continuously, invading surrounding tissue and metastasize to distal organs leading to death. Despite the attack of an immune mechanism or anticancer therapy, cancer cells survive, continually develop, and cell groups that are most suitable for survival selectively propagate. Cancer cells are living bodies with a high degree of viability, which occur by the mutation of a large number of genes. In order that one cell is converted to a cancer cell and developed to a malignant cancer lump that is detectable in clinics, the mutation of a large number of genes must occur. Thus, in order to diagnose and treat cancer at the root, approaches at a gene level are necessary.
Recently, genetic analysis has been actively attempted to diagnose cancer. The simplest typical method is to detect the presence of ABL: BCR fusion genes (the genetic characteristic of leukemia) in blood by PCR. The method has an accuracy rate of more than 95%, and after the diagnosis and therapy of chronic myelocytic leukemia using this simple and easy genetic analysis, this method is being used for the assessment of the result and follow-up study. However, this method has the deficiency that it can be applied only to some blood cancers.
Furthermore, another method has been attempted, in which the presence of genes expressed by cancer cells is detected by RT-PCR and blotting, thereby diagnosing cancer cells present in blood cells. However, this method has shortcomings in that it can be applied only to some cancers, including prostate cancer and melanoma, has a high false positive rate. In addition, it is difficult to standardize detection and reading in this method, and its utility is also limited (Kopreski, M. S. et al., Clin. Cancer Res., 5:1961, 1999; Miyashiro, I. et al., Clin. Chem., 47:505, 2001).
Recently, genetic testing that uses a DNA in serum or blood plasma has been actively attempted. This is a method of detecting a cancer-related gene that is isolated from cancer cells and released into blood and present in the form of a free DNA in serum. It is found that the concentration of DNA in serum is increased by a factor of 5-10 times in actual cancer patients as compared to that of normal persons, and such increased DNA is released mostly from cancer cells. The analysis of cancer-specific gene abnormalities, such as the mutation, deletion and functional loss of oncogenes and tumor-suppressor genes, using such DNAs isolated from cancer cells, allows the diagnosis of cancer. In this effort, there has been an active attempt to diagnose lung cancer, head and neck cancer, breast cancer, colorectal cancer, and liver cancer by examining the promoter methylation of mutated K-Ras oncogenes, p53 tumor-suppressor genes and p16 genes in serum, and the labeling and instability of microsatellite (Chen, X. Q. et al., Clin. Cancer Res., 5:2297, 1999; Esteller, M. et al., Cancer Res., 59:67, 1999; Sanchez-Cespedes, M. et al., Cancer Res., 60:892, 2000; Sozzi, G. et al., Clin. Cancer Res., 5:2689, 1999).
Meanwhile, in samples other than blood, the DNA of cancer cells can also be detected. A method has been attempted in which the presence of cancer cells or oncogenes in sputum or bronchoalveolar lavage of lung cancer patients is detected by a gene or antibody test (Palmisano, W. A. et al., Cancer Res., 60:5954, 2000; Sueoka, E. et al., Cancer Res., 59:1404, 1999). Additionally, other methods of detecting the presence of oncogenes in feces of colon and rectal cancer patients (Ahlquist, D. A. et al., Gastroenterol., 119:1219-27, 2000) and detecting promoter methylation abnormalities in urine and prostate fluid (Goessl, C. et al., Cancer Res., 60:5941, 2000) have been attempted. However, in order to accurately diagnose cancers that cause a large number of gene abnormalities and show various mutations characteristic of each cancer, a method in which a large number of genes are simultaneously analyzed in an accurate and automatic manner is required. However, such a method has not yet been established.
Accordingly, methods of diagnosing cancer by measuring DNA methylation have recently been proposed. When the promoter CpG island of a certain gene is hyper-methylated, the expression of such a gene is silenced. This is interpreted to be a main mechanism by which the function of this gene is lost even when there is no mutation in the protein-coding sequence of the gene in a living body. In addition, this is analyzed as a factor by which the function of a number of tumor-suppressor genes in human cancer is lost. Thus, analysis of the methylation of the promoter CpG island of tumor-suppressor genes is very helpful in cancer research. An active attempt has been made to analyze the methylation of the promoter CpG island by methods such as methylation-specific PCR (hereinafter, referred to as “MSP”) or automatic base sequencing and to use the analysis results for the diagnosis and screening of cancer.
A significant number of diseases are caused by genetic abnormalities, and the most frequent form of genetic abnormality is a change in the coding sequence of a gene. This genetic change is referred to as mutation. When any gene has a mutation, the structure and function of a protein encoded by the gene change, resulting in abnormalities and deletions, and this mutant protein causes disease. However, an abnormality in the expression of a specific gene can cause disease even in the absence of a mutation in the gene. A typical example thereof is methylation in which a methyl group is attached to the transcription regulatory region of a gene, that is, the cytosine base of the promoter CpG islands, and in this case, the expression of the gene is silenced. This is known as epigenetic change. This is transmitted to offspring and results in the loss of the expression of the relevant protein in the same manner as mutation. Most typically, the expression of tumor suppressor genes is silenced by the methylation of promoter CpG islands in cancer cells, resulting in carcinogenesis (Robertson, K. D. et al., Carcinogensis, 21:461, 2000).
For the accurate diagnosis of cancer, it is important to detect not only a mutated gene but also a mechanism by which the mutation of this gene occurs. Previously, studies were conducted focusing on mutations in a coding sequence, i.e., micro-changes, such as point mutations, deletions and insertions, or macroscopic chromosomal abnormalities. However, in recent years, epigenetic changes were reported to be as important as these mutations, and a typical example of the epigenetic changes is the methylation of promoter CpG islands.
In the genomic DNA of mammal cells, there is the fifth base in addition to A, C, G and T, namely, 5-methylcytosine, in which a methyl group is attached to the fifth carbon of the cytosine ring (5-mC). 5-mC is always attached only to the C of a CG dinucleotide (5′-mCG-3′), which is frequently marked CpG. The C of CpG is mostly methylated by attachment with a methyl group. The methylation of this CpG inhibits a repetitive sequence in genomes, such as Alu or transposon, from being expressed. In addition, this CpG is a site where an epigenetic change in mammalian cells appears most often. The 5-mC of this CpG is naturally deaminated to T, and thus, the CpG in mammal genomes shows only 1% of frequency, which is much lower than a normal frequency (¼×¼=6.25%).
Regions in which CpG are exceptionally integrated are known as CpG islands. The CpG islands refer to sites which are 0.2-3 kb in length, and have a C+G content of more than 50% and a CpG ratio of more than 3.75%. There are about 45,000 CpG islands in the human genome, and they are mostly found in promoter regions regulating the expression of genes. Actually, the CpG islands occur in the promoters of housekeeping genes accounting for about 50% of human genes (Cross, S. et al., Curr. Opin. Gene Develop., 5:309, 1995).
In the meantime, in the somatic cells of normal persons, the CpG islands of such housekeeping gene promoter sites are un-methylated, but imprinted genes and the genes on inactivated X chromosomes are methylated such that they are not expressed during development.
During a cancer-causing process, methylation is found in promoter CpG islands, and the restriction on the corresponding gene expression occurs. Particularly, if methylation occurs in the promoter CpG islands of tumor-suppressor genes that regulate cell cycle or apoptosis, restore DNA, are involved in the adhesion of cells and the interaction between cells, and/or suppress cell invasion and metastasis, such methylation blocks the expression and function of such genes in the same manner as the mutations of a coding sequence, thereby promoting the development and progression of cancer. In addition, partial methylation also occurs in the CpG islands according to aging.
An interesting fact is that, in the case of genes whose mutations are attributed to the development of cancer in congenital cancer but do not occur in acquired cancer, the methylation of promoter CpG islands occurs instead of mutation. Typical examples include the promoter methylation of genes, such as acquired renal cancer VHL (von Hippel Lindau), breast cancer BRCA1, colorectal cancer MLH1, and stomach cancer E-CAD. In addition, in about half of all cancers, the promoter methylation of p16 or the mutation of Rb occurs, and the remaining cancers show the mutation of p53 or the promoter methylation of p73, p 14 and the like.
An important fact is that an epigenetic change caused by promoter methylation causes a genetic change (i.e., the mutation of a coding sequence), and the development of cancer is progressed by the combination of such genetic and epigenetic changes. In a MLH1 gene as an example, there is the circumstance in which the function of one allele of the MLH1 gene in colorectal cancer cells is lost due to its mutation or deletion, and the remaining one allele does not function due to promoter methylation. In addition, if the function of MLH1, which is a DNA restoring gene, is lost due to promoter methylation, the occurrence of mutation in other important genes is facilitated to promote the development of cancer.
Most cancers show three common characteristics with respect to CpG, namely, hypermethylation of the promoter CpG islands of tumor-suppressor genes, hypomethylation of the remaining CpG base sites, and an increase in the activity of methylation enzyme, namely, DNA cytosine methyltransferase (DNMT) (Singal, R. & Ginder, G. D., Blood, 93:4059, 1999; Robertson, K. et al., Carcinogensis, 21:461, 2000; Malik, K. & Brown, K. W., Brit. J. Cancer, 83:1583, 2000).
When promoter CpG islands are methylated, the reason why the expression of the corresponding genes is blocked is not clearly established, but is presumed to be because a methyl CpG-binding protein (MECP) or a methyl CpG-binding domain protein (MBD), and histone deacetylase, bind to methylated cytosine, thereby causing a change in the chromatin structure of chromosomes and a change in histone protein.
It is unsettled whether the methylation of promoter CpG islands directly causes the development of cancer or is a secondary change after the development of cancer. However, it is clear that the promoter methylation of tumor-related genes is an important index to cancer, and thus can be used in many applications, including the diagnosis and early detection of cancer, the prediction of the risk of the development of cancer, the prognosis of cancer, follow-up examination after treatment, and the prediction of a response to anticancer therapy. Recently, an attempt to examine the promoter methylation of tumor-related genes in blood, sputum, saliva, feces or urine and to use the examined results for the diagnosis and treatment of various cancers, has been actively conducted (Esteller, M. et al., Cancer Res., 59:67, 1999; Sanchez-Cespedez, M. et al., Cancer Res., 60:892, 2000; Ahlquist, D. A. et al., Gastroenterol., 119:1219, 2000).
In order to maximize the accuracy of cancer diagnosis using promoter methylation, analyze the development of cancer according to each stage and discriminate a change according to cancer and aging, an examination that can accurately analyze the methylation of all the cytosine bases of promoter CpG islands is required. Currently, a standard method for this examination is a bisulfite genome-sequencing method, in which a sample DNA is treated with sodium bisulfite, and all regions of the CpG islands of a target gene to be examined is amplified by PCR, and then, the base sequence of the amplified regions is analyzed. However, this examination has the problem that there are limitations to the number of genes or samples that can be examined at a given time. Other problems are that automation is difficult, and much time and expense are required.
In the Johns Hopkins School of Medicine, the MD Anderson Cancer Center, Charité-Universitatsmedizin Berlin, etc., studies on promoter methylation of cancer-related genes have been actively conducted. The fundamental data thus obtained are interchanged through the DNA Methylation Society (DMS) and stored in MethDB (www.methdb.de). Meanwhile, EpiGenX Pharmaceuticals, Inc. is now developing therapeutic agents associated with the methylation of CpG islands, and Epigenomics, Inc. is now conducting studies to apply promoter methylation to cancer diagnosis by examining the promoter methylation using various techniques, such as DNA chips and MALDI-TOF.
Accordingly, the present inventors have made extensive efforts to develop an effective colorectal-cancer-specific methylation marker which makes it possible to diagnose cancer and the risk of carcinogenesis at an early stage and predict cancer prognosis. As a result, the present inventors have found that SDC2 (NM_002998, Syndecan 2), SIM1 (NM_05068, Single-minded homolog 1 (Drosophila) and SORCS3 (NM_014978, Sortilin-related VPS10 domain containing receptor 3) genes are methylated specifically in colorectal cancer cells and that colorectal cancer can be diagnosed by measuring the degree of methylation using these genes as biomarkers, thereby completing the present invention.